ablaette

library(dplyr)
library(polmineR)
library(data.table)
# library(quanteda)

dict <- c("Migration", "Flucht", "Asyl.*")

speeches <- corpus("GERMAPARL2") %>% 
  subset(protocol_year %in% 2013:2021) %>% 
  as.speeches(s_attribute_date = "protocol_date", s_attribute_name = "speaker_who")

ablaette

library(pdftools)
library(dplyr)
library(tibble)

fname <- "https://www.gruene-bw.de/wp-content/uploads/2021/01/GrueneBW-Landtagswahlprogramm-2021-Wachsen-wir-ueber-uns-hinaus.pdf"
tmp <- tempfile(fileext = ".pdf")
download.file(url = fname, destfile = tmp)

doc <- pdftools::pdf_text(tmp) %>%
  gsub("-\\n\\s*", "", .) %>%

ablaette

# Plot time series for multiple queries

library(polmineR)
library(dplyr)
library(tidyr)
library(xts)
library(lubridate) # we need lubridate::floor_date()

corpus_id <- "GERMAPARL2"
look_up <- "'[Ii]ndustriepolit.*'"

ablaette

# snippet for 'Methodenanwendung in Praxisfeldern':
# Framing immigration as "illegal"

library(polmineR)
library(dplyr)
library(xts)
library(lubridate) # we need lubridate::floor_date()

look_up <- '"([iI]llegal.*|[fF]alsch.*|[uU]ndoku.*)" "(Migr.*|Flüchtl.*|Asyl.*)"'

ablaette

install.packages("remotes")
remotes::install_github("PolMine/cwbtools", ref = "dev")
library(cwbtools)
corpus_install(doi = "10.5281/zenodo.7949074")

ablaette

# last edit: 2023-06-23

library(polmineR)
library(data.table)
library(xts)

afd <- corpus("GERMAPARL2") %>% 
  subset(speaker_party == "AfD") %>% 
  subset(protocol_lp == "19") %>% 
  subset(p_type == "speech")

ablaette

library(polmineR)

afd <- corpus("GERMAPARL2") %>% 
  subset(speaker_party == "AfD") %>% 
  subset(protocol_lp == "19") %>% 
  subset(p_type == "speech")

afd_count <- afd %>%
  count(p_attribute = "word")

ablaette

# Plot time series for multiple queries

library(polmineR)
library(dplyr)
library(tidyr)
library(xts)
library(lubridate) # we need lubridate::floor_date()

corpus_id <- "NADIRATAZ"
look_up <- c("'Einwanderung.*'", "'Zuwanderung.*'")

ablaette

library(topicmodels)
library(dplyr)
library(magrittr)
library(lubridate)
library(data.table)
library(xts)

lda <- readRDS("~/Downloads/germaparl_lda_speeches_250.rds")

topic_to_get <- 133 # Staatsangehörigkeit / Integration

ablaette

# Plot time series for query matches using polmineR/dplyr/xts

library(polmineR)
library(dplyr)
library(xts)
library(lubridate) # we need lubridate::floor_date()

look_up <- '"Arbeitslosigkeit"'

corpus("GERMAPARL") %>%